YourWishes/cglm
1
0
Fork 0
mirror of https://github.com/recp/cglm.git synced 2026-02-17 03:39:05 +00:00
Code Issues Packages Projects Releases Wiki Activity

neon: implement matrix multiplication for ARM Neon

Browse Source
This commit is contained in:
Recep Aslantas
2017-07-11 16:33:41 +03:00
parent 46303a2fb1
commit 3b159cdf02
3 changed files with 74 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
include/cglm-mat.h
View File

@@ -45,8 +45,18 @@
#define cglm_mat_h
#include "cglm-common.h"
#include "arch/simd/cglm-mat-simd-sse2.h"
#include "arch/simd/cglm-mat-simd-avx.h"
#ifdef CGLM_SSE_FP
# include "arch/simd/cglm-mat-simd-sse2.h"
#endif
#ifdef CGLM_AVX_FP
# include "arch/simd/cglm-mat-simd-avx.h"
#endif
#ifdef CGLM_NEON_FP
# include "arch/simd/neon/mat4.h"
#endif
#include <assert.h>
@@ -215,6 +225,8 @@ glm_mat4_mul(mat4 m1, mat4 m2, mat4 dest) {
glm_mat4_mul_avx(m1, m2, dest);
#elif defined( __SSE__ ) || defined( __SSE2__ )
glm_mat4_mul_sse2(m1, m2, dest);
#elif defined( __ARM_NEON_FP )
glm_mat4_mul_neon(m1, m2, dest);
#else
float a00 = m1[0][0], a01 = m1[0][1], a02 = m1[0][2], a03 = m1[0][3],
a10 = m1[1][0], a11 = m1[1][1], a12 = m1[1][2], a13 = m1[1][3],